Deep Multimodal Fusion Autoencoder for Saliency Prediction of RGB-D Images

Figure 3

Some failure cases. (a) RGB. (b) Ground-truth. (c) Proposed.

(a)

(b)

(c)